Health

Ashish Jha on the weak science linking Tylenol and autism

Ashish Jha on the weak science linking Tylenol and autism

The real danger isn’t Tylenol, it’s bad information. The press conference has created confusion and fear, especially among pregnant women. Here are the facts: Long-term high fever and severe pain can present a serious risk during pregnancy for both the mother and the baby, such as birth defects or pre-term labor. One option many of us use to treat fever and pain, ibuprofen (found in Advil and Motrin), has serious and well-known risks in pregnancy, including the possibility of miscarriage. Pregnant women are rightly warned against taking it. This leaves acetaminophen (found in Tylenol) as the only viable option, which has for decades been considered the safest option for pain and fever during pregnancy.
Scaring expectant mothers away from the one medicine that doctors widely agree is safe leaves them with untenable choices. They may suffer through the fever, which itself can be dangerous and lead to neurodevelopmental disorders for their child. They may also turn to alternatives known to cause harm. This is not “playing it safe.” It is reckless advice that may lead to real harm.
In addition, Health and Human Services Secretary Robert F. Kennedy’s insinuation that tylenol causes autism is exceptionally harmful for mothers already raising children with autism, implying that their parenting decisions brought on their child’s condition. That unfounded blame is false and profoundly cruel. Families of children living with autism deserve empathy, support, and recognition for the care, strength, and love they give their children every day.
The science behind this claim doesn’t hold up, and the evidence is being misrepresented. Despite being framed as a new study, the work the administration cited did not generate any new data, and none of it was based on randomized control trials, the gold standard for understanding causal relationships. Instead, it reanalyzed 46 existing observational studies — only seven of which even looked at Tylenol and autism. The rest examined different health outcomes like ADHD.
When researchers evaluate the rigor of observational studies like these, two criteria stand out as critical: how large and representative was the sample of people studied and whether key variables that might influence the results were accounted for. The biggest and best study in this cohort, a Swedish study following 2.4 million children over 25 years, was able to control for several important variables by studying 1,773,747 full siblings. Sibling comparisons allow us to see what happens when one child might have been exposed to Tylenol in the womb while another sibling was not. Because siblings in the study shared many of these same factors, like genetics of both parents, home life, and medical care during pregnancy, comparing them helped researchers test whether Tylenol itself was causing autism. In these studies, any Tylenol links to autism disappeared.
However, instead of putting weight on the strongest results — as is normally done in this kind of research — authors of the systematic review the president and health secretary cited emphasized findings from weaker, less well-done studies. Emphasizing weak associations while downplaying the strongest evidence lent unearned credibility to Kennedy’s pronouncement and gave the White House cover for a claim that simply isn’t supported by science.
Most disappointing of all, the consequences of the administration’s rhetoric fall hardest on those who deserve so much better. By pretending to have discovered the “cause” of autism, HHS and the White House may claim a political victory. But the real losers are pregnant women left anxious about whether to treat pain or fever, and parents with autistic children, made to feel that their own choices caused their child’s condition.
The rise of autism diagnoses over the past several decades has been, and must continue to be, a significant area of research. Much of the rise can be attributed to improved recognition and diagnostic criteria, which ultimately benefits parents and families who can better connect children to the support they need and deserve.
Concrete research is needed to understand the true causes of autism, and what may be driving its increased rate of diagnoses. Further, there needs to be rigorous research on how to best treat children with autism. Instead of focusing energy here, the Trump administration is diverting Americans into a fruitless blame game that stigmatizes families with false narratives and serves them false hope. That is a profound betrayal of the very people the nation’s health agencies are meant to serve. The cost of this distraction is deep and deadly.
Leaders should build trust, not erode it; clarify uncertainty, not exploit it. On Tylenol and autism, one thing is clear: There is no credible evidence that the former causes the latter. What we do know is that fear and misinformation can cause real harm. That is why these claims — and the weak science used to support them — must be called out for what they are: reckless, disappointing, and dangerous.